diff --git a/redfam.py b/redfam.py index 519fe81..2fcddba 100644 --- a/redfam.py +++ b/redfam.py @@ -30,7 +30,9 @@ import locale import re from datetime import datetime +import mwparserfromhell as mwparser # noqa import pywikibot +from pywikibot.tools import deprecated # noqa import jogobot from mysqlred import MysqlRedFam @@ -40,12 +42,12 @@ class RedFam: """ Basic class for RedFams, containing the basic data structure """ - + def __init__( self, articlesList, beginning, ending=None, red_page_id=None, status=0, fam_hash=None, heading=None ): """ Generates a new RedFam object - + @param articlesList list List of articles @param beginning datetime Beginning date @param ending datetime Ending date @@ -54,7 +56,7 @@ class RedFam: @param fam_hash str SHA1 hash of articlesList @param heading str Original heading of RedFam (Link) """ - + # Initial attribute values self._articlesList = articlesList self._beginning = beginning @@ -63,18 +65,18 @@ class RedFam: self._status = status self._fam_hash = fam_hash self._heading = heading - + # Calculates the sha1 hash over self._articlesList to # rediscover known redundance families self.calc_fam_hash() - + def __repr__( self ): """ Returns repression str of RedFam object - + @returns str repr() string """ - + __repr = "RedFam( " + \ "articlesList=" + repr( self._articlesList ) + \ ", heading=" + repr( self._heading ) + \ @@ -84,28 +86,28 @@ class RedFam: ", status=" + repr( self._status ) + \ ", fam_hash=" + repr( self._fam_hash ) + \ " )" - + return __repr def calc_fam_hash( self ): """ Calculates the SHA-1 hash for the articlesList of redundance family. Since we don't need security SHA-1 is just fine. - + @returns str String with the hexadecimal hash digest """ - + h = hashlib.sha1() h.update( str( self._articlesList[:8] ).encode('utf-8') ) - + if self._fam_hash and h.hexdigest() != self._fam_hash: raise RedFamHashError( self._fam_hash, h.hexdigest() ) - + elif self._fam_hash: return else: self._fam_hash = h.hexdigest() - + @classmethod def flush_db_cache( cls ): """ @@ -119,28 +121,28 @@ class RedFamParser( RedFam ): Provides an interface to RedFam for adding/updating redundance families while parsig redundance pages """ - + # Define the timestamp format __timestamp_format = jogobot.config['redundances']['timestamp_format'] - + # Define section heading re.pattern __sectionhead_pat = re.compile( r"^(=+)(.*\[\[.+\]\].*\[\[.+\]\].*)\1" ) # Define timestamp re.pattern __timestamp_pat = re.compile( jogobot.config['redundances'] ['timestamp_regex'] ) - + # Textpattern for recognisation of done-notices __done_notice = ":Archivierung dieses Abschnittes \ wurde gewünscht von:" __done_notice2 = "{{Erledigt|" - + def __init__( self, heading, red_page_id, red_page_archive, beginning, ending=None ): """ Creates a RedFam object based on data collected while parsing red_pages combined with possibly former known data from db - + @param red_fam_heading str Wikitext heading of section @param red_page_id int MediaWiki page_id @param red_page_archive bool Is red_page an archive @@ -149,57 +151,57 @@ class RedFamParser( RedFam ): @param ending datetime Timestamp of ending str strptime parseable string """ - + # Set object attributes: self._red_page_id = red_page_id self._red_page_archive = red_page_archive self._fam_hash = None - + # Method self.add_beginning sets self._beginning directly self.add_beginning( beginning ) - + # Method self.add_ending sets self._ending directly if( ending ): self.add_ending( ending ) else: # If no ending was provided set to None self._ending = None - + self._status = None - + # Parse the provided heading of redundance section # to set self._articlesList self.heading_parser( heading ) - + # Calculates the sha1 hash over self._articlesList to # rediscover known redundance families - + self.calc_fam_hash() - + # Open database connection, ask for data if existing, # otherwise create entry self.__handle_db() - + # Check status changes self.status() - + # Triggers db update if anything changed self.changed() - + def __handle_db( self ): """ Handles opening of db connection """ - + # We need a connection to our mysqldb self.__mysql = MysqlRedFam( ) self.__mysql.get_fam( self._fam_hash ) - + if not self.__mysql.data: self.__mysql.add_fam( self._articlesList, self._heading, self._red_page_id, self._beginning, self._ending ) - + def heading_parser( self, heading ): """ Parses given red_fam_heading string and saves articles list @@ -238,43 +240,43 @@ Maximum number of articles in red_fam exceeded, maximum number is 8, \ def add_beginning( self, beginning ): """ Adds the beginning date of a redundance diskussion to the object - + @param datetime datetime Beginning date """ - + self._beginning = self.__datetime( beginning ) - + def add_ending( self, ending ): """ Adds the ending date of a redundance diskussion to the object. - + @param datetime datetime Ending date """ - + self._ending = self.__datetime( ending ) - + def __datetime( self, timestamp ): """ Decides wether given timestamp is a parseable string or a datetime object and returns a datetime object in both cases - + @param datetime timestamp Datetime object str timestamp Parseable string with timestamp - + @returns datetime Datetime object """ - + # Make sure locale is set to 'de_DE.UTF-8' to prevent problems # with wrong month abreviations in strptime locale.setlocale(locale.LC_ALL, 'de_DE.UTF-8') - + if( isinstance( timestamp, datetime ) ): return timestamp else: result = datetime.strptime( timestamp, type( self ).__timestamp_format ) return result - + def status( self ): """ Handles detection of correct status @@ -284,10 +286,10 @@ Maximum number of articles in red_fam exceeded, maximum number is 8, \ - 2 Discussion archived --> ending (normaly) present, page is archive - 3 and greater status was set by worker script, do not change it """ - + # Do not change stati set by worker script etc. if not self.__mysql.data['status'] > 2: - + # No ending, discussion is running: # Sometimes archived discussions also have no detectable ending if not self._ending and not self._red_page_archive: @@ -299,36 +301,36 @@ Maximum number of articles in red_fam exceeded, maximum number is 8, \ self._status = 2 else: self._status = self.__mysql.data[ 'status' ] - + def changed( self ): """ Checks wether anything has changed and maybe triggers db update """ - + # On archived red_fams do not delete possibly existing ending if( not self._ending and self._status > 1 and self.__mysql.data[ 'ending' ] ): - + self._ending = self.__mysql.data[ 'ending' ] - + # Since status change means something has changed, update database if( self._status != self.__mysql.data[ 'status' ] or self._beginning != self.__mysql.data[ 'beginning' ] or self._ending != self.__mysql.data[ 'ending' ] or self._red_page_id != self.__mysql.data[ 'red_page_id' ] or self._heading != self.__mysql.data[ 'heading' ]): - + self.__mysql.update_fam( self._red_page_id, self._heading, self._beginning, self._ending, self._status ) - + @classmethod def is_sectionheading( cls, line ): """ Checks wether given line is a red_fam section heading - + @param str line String to check - + @returns bool Returns True if it is a section heading """ @@ -336,7 +338,49 @@ Maximum number of articles in red_fam exceeded, maximum number is 8, \ return True else: return False - + + @classmethod + def extract_dates( cls, text, isarchive=False ): + """ + Returns tuple of the first and maybe last timestamp of a section. + Last timestamp is only returned if there is a done notice or param + *isarchiv* is set to 'True' + + @param text Text to search in + @type line Any Type castable to str + @param isarchive If true skip searching done notice (on archivepages) + @type isarchive bool + + @returns Timestamps, otherwise None + @returntype tuple of strs + """ + + # Match all timestamps + matches = cls.__timestamp_pat.findall( str( text ) ) + if matches: + + # First one is beginning + # Since some timestamps are broken we need to reconstruct them + # by regex match groups + beginning = ( matches[0][0] + ", " + matches[0][1] + ". " + + matches[0][2] + ". " + matches[0][3] ) + + # Last one maybe is ending + # Done notice format 1 + # Done notice format 2 + # Or on archivepages + if ( cls.__done_notice in text or + cls.__done_notice2 in text or + isarchive ): + + ending = ( matches[-1][0] + ", " + matches[-1][1] + ". " + + matches[-1][2] + ". " + matches[-1][3] ) + + else: + ending = None + + return (beginning, ending) + @classmethod def is_beginning( cls, line ): """ @@ -402,17 +446,17 @@ class RedFamWorker( RedFam ): where discussion is finished """ def __init__( self, mysql_data ): - + articlesList = [] for key in sorted( mysql_data.keys() ): if 'article' in key and mysql_data[ key ]: articlesList.append( mysql_data[ key ] ) - + super().__init__( articlesList, mysql_data[ 'beginning' ], mysql_data[ 'ending' ], mysql_data[ 'red_page_id' ], mysql_data[ 'status' ], mysql_data[ 'fam_hash' ], mysql_data[ 'heading' ] ) - + @classmethod def list_by_status( cls, status ): """ @@ -426,12 +470,12 @@ class RedFamWorker( RedFam ): print(fam) raise - + class RedFamError( Exception ): """ Base class for all Errors of RedFam-Module """ - + def __init__( self, message=None ): """ Handles Instantiation of RedFamError's @@ -440,12 +484,12 @@ class RedFamError( Exception ): self.message = "An Error occured while executing a RedFam action" else: self.message = message - + def __str__( self ): """ Output of error message """ - + return self.message @@ -453,12 +497,12 @@ class RedFamHashError( RedFamError ): """ Raised when given RedFamHash does not match with calculated """ - + def __init__( self, givenHash, calculatedHash ): - + message = "Given fam_hash ('{given}') does not match with \ calculated ('{calc}'".format( given=givenHash, calc=calculatedHash ) - + super().__init__( message ) @@ -467,8 +511,8 @@ class RedFamHeadingError ( RedFamError ): Raised when given RedFamHeading does not match __sectionhead_pat Regex """ def __init__( self, heading ): - + message = "Error while trying to parse section heading. Given heading \ '{heading}' does not match RegEx".format( heading=heading ) - + super().__init__( message )